OpenAI Introduces Two New Transcription Models

OpenAI announced GPT-Live-Transcribe for low-latency live transcription and GPT-Transcribe for asynchronous file and batch workloads. Both models use context to deliver higher accuracy on real-world audio. Builders can supply free-form context, keywords, expected languages, and prior transcribed turns to refine output. The updates are positioned for developers working with the OpenAI Platform and related tools. Documentation covers choosing between file and realtime transcription modes.